New event, new organiser

[Somnolent]

I'm not convinced that web sites like BikeHike, Bike Route Toaster, etc. are so massively inconsistent to produce figures like "plus or minus 50%"

Maybe not - but no two sites ever give the same answer for the same track.
And third party sites can so easily suffer from what I think of as the "GMW sysndrome" - they can change to what the site owner thinks is a better way of doing things, but it turns out to be less than optimal for some users. 

Given that a single metre difference in elevation can make the difference between AAA points or not, any variation is unhelpful.
This is why AUK is actively pursuing its own tool that will give a consistent result (and save the AAA man a shedload of time into the bargain)
Look for more details in the forthcoming Arrivée.

[frankly frankie]

Elevation figures from sites are always wrong, in a way or another. To measure it accurately, one should ride the course with a decent altimeter, possibly on a day with constant barometric pressure.

Contour-counting an OS 1:50,000 map gives far better results.  It only takes about 20 minutes to contour-count a 200km.

[whosatthewheel]

Elevation figures from sites are always wrong, in a way or another. To measure it accurately, one should ride the course with a decent altimeter, possibly on a day with constant barometric pressure.

Contour-counting an OS 1:50,000 map gives far better results.  It only takes about 20 minutes to contour-count a 200km.

Underestimate though... as you only consider if you go up or down by at least 10 metres... one could argue that going up or down 5 metres is not climbing, but if you do it constantly all day, it can be a long day...

[frankly frankie]

I think I once rode from Doncaster to York without crossing a contour line - and that included riding past a transmitter mast 

5m on average, for each high or low point - on any kind of hilly route it would be an insignificant percentage of the total, and certainly better than an isolated GPS-derived or DEM-derived figure.

[grams]

I've written code that calculates elevation from GPX tracks, using its own topographic database. The problem you quickly run into is that if you add up all the tiny ups and downs for every point you end up with a number that's far too high for flat-ish routes.

So you then have to write some code that filters small undulations* and has some heuristic for what counts as a "climb" (minimum gradient, minimum total height, minimum length, etc). On some routes tiny changes to the chosen values here lead to big differences in the total. So +/- 50% is a totally reasonable number, especially allowing for inaccuracies or lack of detail in the GPX track and the topographic data.

And at some point it comes down to defining what elevation even is. Is riding along a road that constantly undulates slightly the same as riding one that goes up for a long time then down for a long time?

(* note the traditional contour counting method has a similar filter built-in, since small undulations don't cross a contour)

[mattc]

Fractals.

How long is the British coastline?

[Delph Cyclist]

Elevation figures from sites are always wrong, in a way or another. To measure it accurately, one should ride the course with a decent altimeter, possibly on a day with constant barometric pressure.

Contour-counting an OS 1:50,000 map gives far better results.  It only takes about 20 minutes to contour-count a 200km.

Contour-counting certainly has an element of consistency, as two people should at least come up with the same figure for the same ride.

[Somnolent]

Fractals.
How long is the British coastline?

Indeed.  But in practice it doesn't matter.  What's needed is a "good enough" tool that is internally consistent and, as previously mentioned, is something over which AUK has control.

Elevation figures from sites are always wrong, in a way or another. To measure it accurately, one should ride the course with a decent altimeter, possibly on a day with constant barometric pressure.

Contour-counting an OS 1:50,000 map gives far better results.  It only takes about 20 minutes to contour-count a 200km.

Contour-counting certainly has an element of consistency, as two people should at least come up with the same figure for the same ride.

Unfortunately the volume of rides submitted for AAAssessment means that 20 minutes per ride far exceeds the human resource available.
Automated contour counting has been tried but has, so far, failed to produce satisfactory results.

Happily one of our number has come up with a better solution, one that is mostly consistent with values for climbing derived by traditional methods, and takes it a stage or two further by deriving the AAA points available - it even finds and calculates AAA points where only a portion of the ride is sufficiently scenic.

... more details in the forthcoming Arrivée.

[citoyen]

By the way, Bob, I've been putting together the calendar pages for the new Arrivée and have been slightly surprised to note the number of rides listed that don't give a climbing figure, including some put on by very experienced organisers. It's by no means exclusively a 'mistake' of novice organisers, so don't let anyone make you feel bad about it!